Quant from Scratch by QuantWright

Introduction to Quant Finance · Lesson 18

Position sizing

How much on each trade? The same rule with the same trades gives very different accounts, depending on how much each trade risks.

Know the basics? Read the lesson notes

The rule doesn't say how much

A trading rule says when to buy and when to sell. How much each trade buys is a separate choice, called position sizing.

From lesson A7: a stop is the price, set before a trade is opened, at which the trade is sold if it goes wrong. The money lost at the stop is the trade's risk. A result counted in units of that risk is in R: a trade sold at its stop is −1R.

Here is a made-up rule. Each trade ends at −1R or at +1R (a gain as big as its risk), and it wins 55 trades in 100. That is an edge: an advantage that makes money on average. A simulated account starts with €10,000 and risks €100 on every trade. Play its 250 trades.

One account, €100 at risk on every trade (simulated)

Dashed: its €10,000 start.

Trade0 of 250
Total so far0R
Account€10,000

The account won 137 trades and lost 113: 24 more wins than losses, so +24R. At €100 a trade that is 24 × €100 = €2,400, and the account ended at €12,400.

Its deepest fall from a high was a stretch with 11 more losses than wins: 11 × €100 = €1,100, from €11,300 to €10,200.

The same trades at €200

A second account takes the same 250 trades, but risks €200 on every trade.

The first account ended €2,400 up, and its deepest fall was €1,100. What about the second?

A trade's result in euros is its result in R times its risk. Double the risk and every win and every loss doubles: 24 × €200 = €4,800 up, and 11 × €200 = €2,200 for the deepest fall.
The gain does double: 24 × €200 = €4,800. But every loss doubles too, so the 11 more losses than wins now cost 11 × €200 = €2,200.
The fall does double: 11 × €200 = €2,200. But every win doubles too, so the 24 more wins than losses now make 24 × €200 = €4,800.

A fixed amount, or a fixed fraction

The risk per trade multiplies every result, wins and losses alike. So far it was the same number of euros on every trade: a fixed amount.

A second way is to risk the same share of the account every time: a fixed fraction. At 10%, an account that holds €10,000 risks €1,000 on its next trade. Once it holds €9,000, it risks €900.

Two made-up accounts start with €10,000. The first risks a fixed €1,000 on every trade, the second a fixed 10% of what it holds. Both lose ten trades in a row, each at −1R. Where do they end?

Keep your guess in mind. On the next screen you lose the ten trades.

Ten losses in a row

Press the button ten times: ten losses for both accounts.

Two made-up accounts, ten losses in a row

Across: losses in a row. Dark: €1,000 at risk on every trade. Orange: 10% of the account.

Losses in a row0 of 10
€1,000 a trade€10,000
10% a trade€10,000
Its next risk€1,000

The fixed amount lost €1,000 every time, and ten losses took all €10,000.

The fixed fraction lost 10% of what was left each time: €1,000, then €900, then €810, down to €387. It ended at €3,487. Each loss makes the next trade smaller, so by this arithmetic the account never reaches €0.

It is still a large fall: €3,487 is 65% below the start, and getting back takes 6,513 ÷ 3,487 = +187%, because a gain is measured on what is left (lesson A14).

After gains the fixed fraction risks more, not less. Most of the rest of this lesson uses a fixed fraction.

Twice the risk, twice the swing

Back to the 250 trades, now with a fixed fraction: before each trade the account risks the same percent of what it holds. A win adds that percent and a loss takes it.

From lesson A17: an account's drawdown is how far it is below its highest value so far, as a percent of that high. The deepest one is its maximum drawdown.

Slide the risk from 1% to 2%, then to 4%.

The same 250 trades at the risk you set (simulated)

Top: the account (grey: at 1% risk); circles mark its deepest fall. Bottom: its drawdown after each trade.

Ended at€12,555
Maximum drawdown10.5%
Gain to get back to the high+12%

At 1% risk the account ended at €12,555, with a maximum drawdown of 10.5%. At 2%: €15,373 and 20.1%. At 4%: €21,390 and 36.8%.

At 2% every trade moves the account twice as far, in percent, as at 1%, wins and losses alike. So its volatility, the typical size of its moves (lesson A8), doubles.

Here the maximum drawdown grew a little less than in step with the risk, because each loss takes its percent from an account that earlier losses already made smaller. In a longer, choppier fall it can grow more than in step, because a win and a loss of the same percent don't cancel.

The climb back grew faster than the risk. After a 10.5% fall, 89.5% is left, so getting back takes 10.5 ÷ 89.5 = +12%. After 36.8%, 63.2% is left: 36.8 ÷ 63.2 = +58%.

1,000 accounts, three sizes

One list of trades is one run of luck. So the simulation drew 1,000 lists at random from the same rule, which wins 55 trades in 100, and ran an account of €10,000 on each at 1%, 2% and 4% risk. Rank the 1,000 accounts by where they ended: the middle one is the list you played.

Each square is one account. Grey: it never fell 20% below its high. Orange: at some point it was 20% or more below. Dark: 50% or more. Switch between the three sizes.

1,000 accounts on the same rule (simulated)

Sorted, deepest drawdown first.

Met a drawdown of 20% or more36 of 1,000
Of 50% or more0 of 1,000
Middle ending€12,555
100th lowest ending€10,279

At 1% risk, 36 of the 1,000 accounts met a drawdown of 20% or more, and none met one of 50%. At 4%, 988 met one of 20% or more, and 163 were at some point down by half.

The middle account ended above its €10,000 start at each size, because this rule was built with an edge. The 100th lowest, a tenth of the way up from the bottom, ended at €10,279 at 1% risk but at €9,607 at 4%, below its start.

From a risk to a number of shares

An order to buy needs a number of shares (small pieces of a company), not a risk in euros. The stop turns one into the other.

An account of €10,000 risks 1% on its next trade: €100. It buys at €50 a share. With the stop at €48, each share loses €2 if the trade is sold there, so €100 of risk pays for 100 ÷ 2 = 50 shares.

Move the stop until the trade buys exactly 25 shares.

One trade, €100 at risk (made-up prices)

Across: shares. Up: the price. The shaded box is the risk, shares × the fall to the stop: always €100.

Each share loses€2.00
Shares100 ÷ 2 = 50
The shares cost50 × €50 = €2,500
Risk€100

With the stop at €46, each share loses €4, so the trade buys 100 ÷ 4 = 25 shares. They cost 25 × €50 = €1,250: the trade's position size.

With the stop at €49.50 the same €100 of risk buys 200 shares for €10,000, the whole account.

So position size and risk are different numbers: what the shares cost, and what the trade loses at its stop. A closer stop means more shares for the same risk.

Theory

Fixed amount and fixed fraction

Risking the same share of the account on every trade is called fixed-fractional position sizing. With a fixed amount, results add up. With a fixed fraction each result is measured on what the account holds then, so results multiply: compounding (lesson A14).

fixed amount: Vt = Vt−1 + A × Rt
fixed fraction: Vt = Vt−1 × (1 + f × Rt)
Vt
the account after trade t, with no money added or taken out; V0 is its starting value
Vt−1
the account before trade t
Rt
the result of trade t in R: −1 for a trade sold at its stop, +2 for one that made twice its risk
A
the fixed amount: the money risked on every trade
f
the fixed fraction: the share of the account risked on every trade, as a decimal: 2% is 0.02
Worked example: your first three trades

The 250 trades began −1R, −1R, +1R. At a fixed 2%, f = 0.02: V1 = 10,000 × (1 − 0.02) = 9,800. V2 = 9,800 × 0.98 = 9,604. V3 = 9,604 × 1.02 = €9,796.08. At a fixed €200: 10,000 − 200 − 200 + 200 = €9,800.

Holds when
  • each trade's risk is set before the trade: the fixed fraction of what the account holds then, or the fixed amount, which the account must still hold;
  • a losing trade loses its risk and no more: it really is sold at its stop;
  • trades come one at a time.
Breaks when

A price jumps past the stop: a loss bigger than 1R (lesson A13, Orders: market, limit and stop). Trading costs add to every loss (lesson A16, Costs).

New numbers. An account of €8,000 risks a fixed 5% on every trade. Its next three trades end at −1R, −1R and +2R. What does it hold after them, in euros?

8,000 × 0.95 = 7,600. Then 7,600 × 0.95 = 7,220. The +2R trade adds 2 × 5% = 10%: 7,220 × 1.10 = €7,942. The results add up to 0R, but the win was measured on a smaller account.
8,000 × 0.95 = 7,600. Then 7,600 × 0.95 = 7,220. The +2R trade adds 2 × 5% = 10%: 7,220 × 1.10 = €7,942. If you got 8,000, you added the three results up to 0R. That works for a fixed amount of €400. With a fixed fraction the win was measured on a smaller account.
Theory

From risk to shares

Working out the number of shares from the share of the account at risk is often called percent-risk position sizing. For a buy with its stop below the buy price:

N = (f × V) ÷ (Pbuy − Pstop)
position size = N × Pbuy
N
the number of shares to buy, rounded down to a whole number so the trade never risks more than f × V
V
the account before the trade
f
the share of the account to risk on the trade, as a decimal: 1% is 0.01
f × V
the trade's risk in money
Pbuy
the price paid for one share
Pstop
the stop, below the buy price: the price at which the trade is sold if it goes wrong
Pbuy − Pstop
what one share loses if the trade is sold at its stop, with no gap and no cost
Worked example: your trade

V = 10,000 and f = 0.01, so f × V = €100. Buy at €50, stop at €46: N = 100 ÷ (50 − 46) = 100 ÷ 4 = 25 shares. Position size = 25 × €50 = €1,250.

Holds when
  • the stop is set before the buy, and the trade really is sold there;
  • the shares cost no more than the account holds.
Breaks when

A price jumps past the stop (lesson A13, Orders: market, limit and stop): each share loses more than planned. With 200 shares, €1 past the stop adds €200 to a planned €100.

New numbers. An account of €8,000 risks 1.5% on a trade that buys at €20 with its stop at €19.20. How many shares does it buy?

6 is the risk divided by the price: 120 ÷ 20. Divide by what one share loses at the stop instead: 20 − 19.20 = €0.80, and 120 ÷ 0.80 = 150 shares.
The risk is 1.5% of 8,000: €120. One share loses 20 − 19.20 = €0.80 at the stop. N = 120 ÷ 0.80 = 150 shares.
400 shares is the whole account: 8,000 ÷ 20. At the stop they would lose 400 × €0.80 = €320, not the €120 the account set out to risk. 120 ÷ 0.80 = 150 shares.

What this means for your backtest

A backtest runs a trading rule over past prices to see what it would have made. Two things follow. Turn them over.

  1. A result means little without its size.

    "Up 114% with a maximum drawdown of 36.8%" and "up 26% with 10.5%" were the same 250 trades, at 4% and at 1% risk. To compare two backtests, find the risk per trade each used, or count both in R (lesson A7).

Practice 1 of 3

How many shares is that?

The situationA friend describes a plan on paper: an account of €25,000, with 2% of it at risk on each trade. The next trade would buy at €40 a share, with its stop at €37.50.

How many shares does that plan come to?

Hint: the risk in euros, f × V, divided by what one share loses at the stop, Pbuy − Pstop.

The risk is 2% of 25,000: €500. One share loses 40 − 37.50 = €2.50 at the stop. 500 ÷ 2.50 = 200 shares.
The risk is 2% of 25,000: €500. One share loses 40 − 37.50 = €2.50 at the stop. 500 ÷ 2.50 = 200 shares. If you got 12.5, you divided the risk by the price. If you got 625, you spent the whole account.
Practice 2 of 3

"Same gain, a quarter of the drawdown"

The situationA made-up seller writes: "Last year: up 80% with a maximum drawdown of 40%, risking 4% of the account per trade. Too bumpy? Run it at 1%: the same 80%, with a quarter of the drawdown."

Which part of that last sentence can't be right?

The drawdown does shrink: at a quarter of the risk every loss moves the account a quarter as far. But so does every win, so the gain can't stay at 80%.
Size multiplies wins and losses alike. At a quarter of the risk every trade moves the account a quarter as far, so the gain shrinks along with the drawdown.
At a quarter of the risk every trade moves the account a quarter as far, wins as well as losses. In this lesson's simulation, going from 4% risk to 1% took the account from up 114% to up 26%.
Practice 3 of 3

The same backtest at three times the risk

The situationYour own research: a backtest that risked a fixed 1% of the account on each trade shows a maximum drawdown of 12%. You wonder what the same trades would have shown at 3%.

About how deep would the maximum drawdown have been, and what gain would undo it?

Three times the risk moves every trade three times as far: about 3 × 12 = 36%. The exact figure depends on the path: in this lesson's simulation 3% risk gave 28.9% against 10.5% at 1%, a little under three times. After a 36% fall, 64% is left, and 36 ÷ 64 = +56%.
The depth is right. Three times the risk moves every trade three times as far: about 3 × 12 = 36%. The exact figure depends on the path: in this lesson's simulation 3% risk gave 28.9% against 10.5% at 1%, a little under three times. But a gain is measured on what is left. After a 36% fall, 64% is left, and 36 ÷ 64 = +56%.
The trades are the same, but the size is not. Three times the risk moves every trade three times as far: about 3 × 12 = 36%. The exact figure depends on the path: in this lesson's simulation 3% risk gave 28.9% against 10.5% at 1%, a little under three times. After a 36% fall, 64% is left, and 36 ÷ 64 = +56%.
✓

Lesson done

Position sizing sets how much each trade risks, and that risk multiplies every result, wins and losses alike. In this lesson's simulation the same 250 trades ended up 26% with a 10.5% maximum drawdown at 1% risk, and up 114% with 36.8% at 4%. The lesson shows what a size does, not which size to use.

Key ideas

Position sizing
deciding how much each trade buys, and so how much it risks
Fixed amount
risking the same money on every trade
Fixed fraction (fixed-fractional sizing)
risking the same share of the account on every trade, so the money at risk shrinks after losses and grows after gains
Position size
what a trade's shares cost: shares × the price paid for one
Percent-risk position sizing
choosing the number of shares so that a sale at the stop loses a set share of the account

Formulas

fixed amount: Vt = Vt−1 + A × Rt  ·  fixed fraction: Vt = Vt−1 × (1 + f × Rt)

N = (f × V) ÷ (Pbuy − Pstop)  ·  position size = N × Pbuy

Account formulas: results in R, no money added or taken out. Share formula: a buy with its stop below the buy price, sold at the stop with no gap or cost, N rounded down.

Lesson notes

Position sizing is the choice of how much each trade risks. (A trade's position size is a different thing: what its shares cost.) That risk multiplies every result, wins and losses alike, so the same trades give very different accounts at different sizes. These notes show what a size does, not which size to use.

The same trades at two sizes

A stop is the price, set before a trade is opened, at which the trade is sold if it goes wrong. The money lost there is the trade's risk, and a result counted in units of that risk is in R: a trade sold at its stop is −1R (lesson A7).

The lesson's made-up rule has trades that end at −1R or at +1R, and it wins 55 in 100: an edge, an advantage that makes money on average. One simulated list of 250 trades held 137 wins and 113 losses: 24 more wins than losses, so +24R. An account of €10,000 that risked €100 on every trade ended 24 × €100 = €2,400 up. Its deepest fall from a high was a stretch with 11 more losses than wins: 11 × €100 = €1,100.

At €200 a trade the same list ends 24 × €200 = €4,800 up, with a deepest fall of 11 × €200 = €2,200.

A fixed amount or a fixed fraction

Risking the same money on every trade is a fixed amount. Risking the same share of the account every time is a fixed fraction: the money at risk shrinks after losses and grows after gains.

Take two made-up accounts of €10,000 and ten losses in a row at −1R. A fixed €1,000 a trade leaves €0. A fixed 10% loses €1,000, €900, €810 and so on down to €387, and leaves €3,487. Each loss makes the next trade smaller, so by this arithmetic the account never reaches €0.

It is still a large fall: €3,487 is 65% below the start, and getting back takes 6,513 ÷ 3,487 = +187%, because a gain is measured on what is left (lesson A14).

Risk and drawdown

An account's drawdown is how far it is below its highest value so far, as a percent of that high. The deepest one is its maximum drawdown (lesson A17).

The same 250 trades were run with a fixed fraction. At 1% risk the account ended at €12,555, up 26%, with a maximum drawdown of 10.5%. At 2%: €15,373 and 20.1%. At 4%: €21,390, up 114%, and 36.8%. At 2% every trade moves the account twice as far, in percent, as at 1%, so its volatility, the typical size of its moves (lesson A8), doubles.

Here the maximum drawdown grew a little less than in step with the risk, because each loss takes its percent from an account that earlier losses already made smaller. In a longer, choppier fall it can grow more than in step, because a win and a loss of the same percent don't cancel. The climb back grew faster than the risk: after a 36.8% fall, 63.2% is left, so getting back takes 36.8 ÷ 63.2 = +58%.

A backtest runs a trading rule over past prices to see what it would have made. Its result means little without its size: "up 114% with a maximum drawdown of 36.8%" and "up 26% with 10.5%" were the same trades. To compare two backtests, find the risk per trade each used, or count both in R.

1,000 accounts, and then a rule with no edge

One list of trades is one run of luck, so the simulation drew 1,000 lists of 250 trades at random from the same rule and ran each at the three sizes. At 1% risk, 36 of the 1,000 accounts met a drawdown of 20% or more, and none met one of 50%. At 4%, 988 met one of 20% or more, and 163 were at some point down by half.

Rank the 1,000 accounts by their ending value at each size. The middle one in that ranking is always the list above, and it ended above its €10,000 start at each size, because the rule was built with an edge. The 100th lowest, a tenth of the way up from the bottom, ended at €10,279 at 1% risk but at €9,607 at 4%, below its start.

Size can't add an edge. The simulation also ran the three sizes on coin flips: 50 wins in 100, so no edge. There the middle account's list held 125 wins and 125 losses. A win and a loss of the same percent don't cancel (at 4%, 1.04 × 0.96 = 0.9984), so it ended at €9,876 at 1% risk and at €8,186 at 4%: it lost at each size, and the bigger size lost more.

From a risk to a number of shares

An order to buy needs a number of shares (small pieces of a company), and the stop turns a risk into shares. An account of €10,000 that risks 1% has €100 at risk. If it buys at €50 with its stop at €46, each share loses €4 at the stop, so it buys 100 ÷ 4 = 25 shares. They cost 25 × €50 = €1,250: the trade's position size.

With the stop at €49.50 the same €100 of risk buys 200 shares for €10,000, the whole account. Position size is what the shares cost; risk is what the trade loses at its stop. A closer stop means more shares for the same risk.

Formula: fixed amount and fixed fraction

Risking the same share of the account on every trade is called fixed-fractional position sizing. With a fixed amount, results add up. With a fixed fraction each result is measured on what the account holds then, so results multiply: compounding (lesson A14).

fixed amount: Vt = Vt−1 + A × Rt
fixed fraction: Vt = Vt−1 × (1 + f × Rt)
Vt
the account after trade t, with no money added or taken out; V0 is its starting value
Vt−1
the account before trade t
Rt
the result of trade t in R: −1 for a trade sold at its stop, +2 for one that made twice its risk
A
the fixed amount: the money risked on every trade
f
the fixed fraction: the share of the account risked on every trade, as a decimal: 2% is 0.02
Worked example: the first three trades

The 250 trades began −1R, −1R, +1R. At a fixed 2%, f = 0.02: V1 = 10,000 × (1 − 0.02) = 9,800. V2 = 9,800 × 0.98 = 9,604. V3 = 9,604 × 1.02 = €9,796.08. At a fixed €200: 10,000 − 200 − 200 + 200 = €9,800.

Holds whenBreaks when

A price jumps past the stop: a loss bigger than 1R (lesson A13, Orders: market, limit and stop). Trading costs add to every loss (lesson A16, Costs).

Formula: from risk to shares

Working out the number of shares from the share of the account at risk is often called percent-risk position sizing. The formula is for a buy with its stop below the buy price.

N = (f × V) ÷ (Pbuy − Pstop)
position size = N × Pbuy
N
the number of shares to buy, rounded down to a whole number so the trade never risks more than f × V
V
the account before the trade
f
the share of the account to risk on the trade, as a decimal: 1% is 0.01
f × V
the trade's risk in money
Pbuy
the price paid for one share
Pstop
the stop, below the buy price: the price at which the trade is sold if it goes wrong
Pbuy − Pstop
what one share loses if the trade is sold at its stop, with no gap and no cost
Worked example: the trade above

V = 10,000 and f = 0.01, so f × V = €100. Buy at €50, stop at €46: N = 100 ÷ (50 − 46) = 100 ÷ 4 = 25 shares. Position size = 25 × €50 = €1,250.

Holds whenBreaks when

A price jumps past the stop (lesson A13, Orders: market, limit and stop): each share loses more than planned. With 200 shares, €1 past the stop adds €200 to a planned €100.

One more worked example

Made-up round numbers

An account of €20,000 risks a fixed 1.5% on every trade.

The first trade's risk: 0.015 × 20,000 = €300. It buys at €80 with its stop at €77, so each share loses €3 at the stop: N = 300 ÷ 3 = 100 shares, a position size of 100 × €80 = €8,000.

It ends at −1R: 20,000 × (1 − 0.015) = €19,700. The next trade ends at +2R: 19,700 × 1.03 = €20,291. A third ends at −1R: 20,291 × 0.985 = about €19,987.

The three results (−1R, +2R, −1R) add up to 0R, yet the account is about €13 below its start. At a fixed €300 a trade it would be back at exactly €20,000.

Common mistakes

Check yourself

  1. A list of trades adds up to +30R, and its deepest fall is 12R. An account risks a fixed €50 on every trade. What does it gain, and how big is its deepest fall?
    Answer

    It gains 30 × €50 = €1,500. Its deepest fall is 12 × €50 = €600.

  2. An account of €5,000 risks a fixed 2% on every trade and loses two trades in a row at −1R. What does it hold?
    Answer

    5,000 × 0.98 = 4,900, and 4,900 × 0.98 = €4,802. A fixed €100 a trade would leave €4,800.

  3. An account of €12,000 risks 1% on a trade that buys at €30 with its stop at €28.50. How many shares does it buy, and what is the position size?
    Answer

    The risk is €120 and one share loses €1.50 at the stop: 120 ÷ 1.50 = 80 shares. The position size is 80 × €30 = €2,400.

  4. The same trades are run at a fixed 1% and at a fixed 2% of the account. What happens to the account's volatility and to its maximum drawdown?
    Answer

    Every trade moves the account twice as far in percent, so the volatility doubles. The maximum drawdown comes out about twice as deep: in the lesson's simulation it was 10.5% and 20.1%, a little under twice.

  5. (from lesson A14) An account is 36.8% below its high. What gain gets it back?
    Answer

    A gain is measured on what is left. 63.2% is left, so it takes 36.8 ÷ 63.2 = +58%.

Try it yourself

Run the script. Every number in this lesson comes from one Python script with a fixed random seed (below). Change the chance of a win, the sizes or the number of trades and see what moves.

By hand, in a spreadsheet. Type a list of results in R down column A, starting in A2: 1 for a win, -1 for a loss. Put 10000 in B1 and in C1. In B2 put =B1+100*A2, a fixed €100 a trade. In C2 put =C1*(1+0.01*A2), a fixed 1% a trade. Copy both down, then change the 100 and the 0.01. For the shares of one trade: =ROUNDDOWN(0.01*10000/(50-46),0).

The code behind every number (Python, seed 2026111014)
Loading…

Going further: the same sizes with no edge

The script also gave the 1,000 accounts a coin flip in place of the rule: 50 wins in 100, so no edge. The middle account ended at €9,876 at 1% risk, at €9,512 at 2% and at €8,186 at 4%. Its list held 125 wins and 125 losses.

The reason is the volatility drag of lesson A14. A win and a loss of the same percent multiply to a little less than 1: 1.01 × 0.99 = 0.9999, 1.02 × 0.98 = 0.9996 and 1.04 × 0.96 = 0.9984. Twice the risk costs four times as much for each such pair, and 125 pairs at 4% leave 0.9984 multiplied by itself 125 times, which is about 0.8186 of the start.

The falls grew as before. With no edge, 316 of the 1,000 accounts met a drawdown of 20% or more at 1% risk. At 4% all 1,000 did, and 619 were at some point down by half. The number of accounts that ended below their start went from 533 at 1% to 638 at 4%.

The same drag is at work when a rule has an edge. In a model like this one, past some size it outweighs the edge, and a still bigger size makes the middle account end lower. Lesson C10, Kelly, is about that point.

For the preciseSeed 2026111014, one numpy Generator. First draw: 1,000 lists × 250 trades, a win (+1R) with chance 0.55, else a loss (−1R). Second draw: the same on a fair coin. The list you played is "the middle list", row 18: its 137 wins are the middle number of wins among the 1,000 lists (the 500th when sorted), and it is the lowest row with that many (53 lists share it). An account's end depends only on how many of its trades won, so this list is also the 500th lowest ending at every size. The 100th lowest ending belongs to a list with 127 wins and the 100th highest to one with 147. The 100th highest ended at €15,334 at 1% risk, €22,936 at 2% and €47,625 at 4%; the 100th lowest at €10,279, €10,305 and €9,607. The list's lowest total was −7R, early on. The list's deepest fall ran from trade 179 to trade 202: 23 trades with 17 losses and 6 wins. At every size on the slider the maximum drawdown came from those trades; at 2% it was 1.91 times the 1% drawdown, and at 4% it was 3.5 times. With a fixed €100 a trade the same fall, €1,100 from €11,300, is a drawdown of 9.7%. Across the 1,000 accounts the middle maximum drawdown was 10.6% at 1% risk, 20.2% at 2% and 37.3% at 4%, and the deepest of all were 34.3%, 57.8% and 83.7%. Accounts that ended below their start: 75, 92 and 112. At 2% risk, 524 accounts met a drawdown of 20% or more and 4 one of 50% or more. Drawdowns are rounded to 0.01% before they are counted against 20% and 50%.

Honest notes

Where this comes from